Cheng Tan

Assistant Professor · Northeastern University · New York University

⭐ If you are interested in building systems, we should talk [c.tan (at) northeastern.edu].

Research interests

I’m broadly interested in computer systems. I build systems that enable users to verify their outsourced services. In particular, my recent research includes: building Faithful LLM Systems, verifying ML systems, verifying consistency models of concurrent systems, verifying the execution of applications on untrusted servers.

Selected publications (since 2023)

[2026] LLM Jailbreak Oracle - MLSys 2026 [2026] Virtual Machine NUMA Placement at Scale: Learning the Norm, Shielding the Tail - MLSys 2026 [2025] TrainVerify: Equivalence-Based Verification for Distributed LLM Training - SOSP 2025 [2025] PipeLLM: Fast and Confidential Large Language Model Services with Speculative Pipelined Encryption - ASPLOS 2025 [2024] NN4SysBench: Characterizing Neural Network Verification for Computer Systems - NeurIPS 2024 [2024] Dropout Attacks - S&P 2024 [2023] Encrypted Databases Made Secure Yet Maintainable - OSDI 2023 [2023] Building Verified Neural Networks for Computer Systems with Ouroboros - MLSys 2023 [2023] Viper: A Fast Snapshot Isolation Checker - EuroSys 2023 [2023] NNSmith: Generating Diverse and Valid Test Cases for Deep Learning Compilers - ASPLOS 2

Faculty homepage More professors at New York University

Is Cheng Tan a match for your research?

Upload your CV and get matched to professors who fit — with hiring signals.

Start matching